<i>Illustration: Glen Le Lievre.</i>

Illustration: Glen Le Lievre.

The consumerist question about this week's budget is: how did it affect my pocket? The egalitarian question is: was its treatment of people at the bottom, middle and top reasonably fair? But the macro-economic question is: how will the budget affect the economy?

We know the economy has been, and is expected to continue, growing at below its medium-term trend rate of about 3 per cent a year, the rate that keeps unemployment steady. So will the budget help to speed things up or slow them down? In the economists' jargon, will its effect be ''expansionary'' or ''contractionary''?

It may seem a simple question, but economists have various ways of attempting to answer it. One outfit asking itself this question is the Reserve Bank. The Reserve will take account of the budget's effect - along with various other factors' effects - on the strength of demand in the economy in making its monthly decisions about whether to raise, lower or leave unchanged the instrument it uses to affect the strength of demand, the official interest rate.

In making that assessment the Reserve takes a very simple approach: in what direction is the budget balance expected to change between the present financial year and the coming financial year that starts in July? And having determined the direction of the change, how big is it? Obviously, the bigger it is, the more notice we should take of it.

Taken at face value, the answers to those questions aren't ones most people would be pleased to hear. Joe Hockey is expecting a budget deficit of $49.9 billion in the financial year just ending and a deficit of $29.8 billion in the coming year.

That's an expected improvement of $20.1 billion - which may please those people who think getting the government's deficits and debt down as quickly as possible is the only thing that matters, but would worry most business people and economists.

Why? When governments spend more in the economy than they take out of it in tax collections - that is, run a deficit - they're contributing to the net demand for the production of goods and services that keeps the economy growing and increasing employment opportunities. Which, when private demand is weak, is a good thing.

(It would be a different matter if private demand were strong and the additional demand from the public sector was adding to inflation pressure.)

So the expected reduction of $20.1 billion in the budget's net addition to demand will have a contractionary effect which, taken by itself, will tend to make the economy grow even more slowly. And since the budget papers imply nominal gross domestic product will be $1632 billion in 2014-15, a $20.1 billion change represents 1.2 per cent of GDP - making it highly significant.

Oh dear. Doesn't sound good. But, as I say, this is taking the budget figures at face value - always unwise in economics. What's more, simply focusing on the direction and size of the expected change in the budget balance is a bit simplistic.

For a start, Hockey inflated the old year's deficit by choosing to make a payment of $8.8 billion to the Reserve Bank. This is just the government moving money between its pockets; it has no effect on demand.

If you ignore the one-off payment to the Reserve, the expected improvement in the budget deficit falls to $11.3 billion, which is equivalent to 0.7 per cent of GDP - but that's still a quite significant degree of contraction.

But here's where we start getting tricky. When you imagine that reducing the budget deficit by $1 will therefore reduce nominal GDP by $1, you're implicitly assuming that whatever the government does to bring that $1 reduction about won't have any effect on the behaviour of people who've had their benefits cut or their tax increased.

In the economists' jargon, you're assuming a ''multiplier'' of 1. In 2009, however, the Organisation for Economic Co-operation and Development published estimates of the multiplier effects of changes in various classes of government spending and taxation by the Australian government.

It found, for instance, that increased government spending on building new infrastructure would have a multiplier of 0.9 in the first year (and 1.3 in the second year, as the increased spending by the government prompted the eventual recipients of that money to increase their own spending).

By contrast, it found that, on average, an increase in government spending on ''transfers to households'' (such as a cash splash) had a multiplier of just 0.4 in the first year, rising to 0.8 in the second year.

Why? Because a lot of people would hang on to the money (save it, or use it to reduce their debts) rather than spend it, particularly at first.

This explains why the OECD's multiplier for a cut in income tax is only 0.4 - people would save most of it. Similarly, an increase in income tax would reduce consumer spending by only 60 per cent of the increase because some people would cut their rate of saving to ''smooth'' their consumption.

The OECD's various multipliers for Australia range from 0.3 to 1.3. If we use a narrower range closer to the middle of that range - 0.6 to 0.9 - and apply these multipliers to the 0.7 per cent of GDP we calculated earlier, we get an estimated negative impact on GDP of between 0.4 and 0.6.

This suggests the budget's negative effect on demand won't be too terrible.

And note this: most of the expected improvement in the deficit in 2014-15 comes from an expected improvement in the economy (more people paying more tax; fewer people needing assistance) rather than from all the tough changes Hockey announced on Tuesday night.

The lion's share of the budget savings don't come until 2017-18. Why? Partly for political reasons but also because, as he's long been saying, Hockey didn't want to hit the economy while it was down.

Ross Gittins is the economics editor.